Every captured commit receives deterministic security triage and a separate communication-quality score. Security candidates and broader second-pass signals receive full-patch Ollama analysis.
Message quality measures whether a commit identifies its scope, purpose, rationale, testing, and supporting references. It does not change the security-severity score.
68/100 average clarity
702Strong · 80–100
1135Adequate · 60–79
582Thin · 40–59
151Opaque · 0–39
20security candidates with opaque commit messaging
This is a routine code cleanup in a Bitcoin library. It renames a simple data container from KeySource to KeyOrigin and gives it named fields (fingerprint and path) instead of numbered tuple slots (.0 and .1). The serialized format is inte…
This change is a routine internal cleanup: the key_expression crate stops depending directly on the secp256k1 cryptography library and instead uses a re-export provided by another internal crate called crypto. No user-facing behavior, type…
This is a small Cargo.toml fix in a Rust Bitcoin subcrate. When the optional 'serde' feature was enabled on the key_expression crate alone, it did not also enable serde on the network crate. That meant a type called NetworkKind could not b…
This is a routine update to a GitHub Actions helper used in automated testing workflows. It bumps the version of taiki-e/install-action, which downloads and installs Rust developer tools during CI runs. There is no change to the actual Bit…
This is a routine update to a GitHub Actions helper used only in automated security-scanning workflows. The new version adds minor improvements, including respecting proxy settings and verifying downloaded files with checksums. It does not…
Dependency bump of a third-party GitHub Action used in CI onlyUpstream release notes mention generic security improvements (checksum verification, npm install defaults, NO_PROXY)No CVE, advisory, or vendor security disclosure is present in the commit or supplied references
This commit removes automatic serde (serialization/deserialization) code generation from several Rust Bitcoin library crates and replaces it with hand-written implementations. The stated goal is to reduce dependencies and make the code eas…
Manual serde implementations replace derived ones, increasing the surface area for serialization-format bugsNo validation logic changes are visible in the diffNo mention of security fixes, CVEs, or vulnerability reports in commit message or PR description
This commit updates the project's automated Rust compiler version from 1.99.1 to 1.100.0 and adjusts a CI cache cleanup script to match a new directory layout used by Cargo 1.100. It is purely an infrastructure/maintenance change to keep a…
This is a build-system improvement, not a code change. It updates a CI helper tool and makes the automated API-checking job fail if a private internal crate accidentally becomes part of the public API. That prevents future leaks of interna…
CI guardrail added to prevent private dependency leaks into stable crate public APIsTool bump enables detection of re-exported private dependenciesNo source-code vulnerability is patched
This is a minor internal code cleanup in the Rust Bitcoin library. It changes how two functions (Xpriv::decode and Xpub::decode) build their error messages when an input is the wrong length. The behavior is the same before and after; only …
This is a routine API cleanup, not a security fix. The change removes one way of naming a Bitcoin unit ('MicroBitcoin' / 'uBTC') and keeps the equivalent 'bits' form. Any code that previously used the old name will fail to compile or fail …
No security signal: public API deprecation and enum variant removalNo memory safety, cryptographic, or consensus code modifiedError message change is informational only
This commit is a code cleanup and feature-gating refactor for the rust-bitcoin library's key_expression crate. It makes more BIP32 (Bitcoin wallet key derivation) types available when the 'alloc' memory allocator feature is disabled, and s…
Refactor of error types and parsing paths for Xpriv/Xpub base58 decodingSwitch from unbounded base58 decode to fixed-size array decode (78 bytes)Removal of explicit length-check error path in favor of decoder-reported length error
This change only adds `#[cfg_attr(miri, ignore)]` annotations to a handful of slow unit tests so they are skipped when run under Miri, a Rust interpreter used for detecting undefined behavior. It does not change any production code, consen…
This commit only updates version numbers in package configuration files (Cargo.toml) for internal dependencies so that all crates in the project require the same minimum versions of shared libraries. There are no code changes, no bug fixes…
This is a routine update to a GitHub Actions helper used in automated testing workflows. It changes which version of a third-party tool-installer action is used, but only for installing the cargo-mutants testing tool. There is no indicatio…
This commit updates the version of a helper tool called cargo-binstall used in two automated GitHub workflows. The change only affects internal build/CI tooling and does not alter the rust-bitcoin library code that users install or run. Th…
This is a routine update by Dependabot to the GitHub CodeQL upload-sarif action used in a scheduled security scanning workflow. It changes one pinned commit hash to a newer version of the same action. There is no indication this introduces…
This commit is a minor code cleanup in the Rust Bitcoin library. It changes how an error message is built when decoding BIP32 extended keys, switching from one nested construction style to two chained calls. The actual behavior and returne…
This commit fixes a bug in how optional Bitcoin amount and fee-rate values are read from JSON when those fields are inside a flattened struct and the value is null. Previously, the code would error out because it did not handle a special s…
Deserialization failure in optional numeric types under serde flattenMissing serde visitor method (visit_unit) causing runtime errorsRegression test added for null handling in flattened structs
This is a routine dependency update by Dependabot, bumping the version of a GitHub-provided action used to upload security scan results. There is no indication of a security vulnerability in the project itself or in this change.
This is a routine automated update by Dependabot that changes the version of a GitHub Action used to install Rust tools in two CI workflows. There is no indication of a security issue in the commit itself, and no verified security referenc…
Expand any commit for its author, full message, clarity score, changed files, triage signals, analysis, and source link.
Security candidateMerge rust-bitcoin/rust-bitcoin#7005: feat(units): add saturating arithmetic methods to FeeRateby Andrew Poelstra · 9079f0cb · Oct 11, 2026 · 1 fileMessage 96 · StrongInformational 18Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7005: feat(units): add saturating arithmetic methods to FeeRate
ab56fd63b79c22966bc35edda724d8fdbadc6860 feat(units): add saturating arithmetic methods to FeeRate (luhrhenz)
Pull request description:
## Summary
- Add `saturating_mul`, `saturating_div`, `saturating_add`, and `saturating_sub` to `FeeRate`, following the pattern established by `BlockHeight` and the existing `checked_*` methods. - All four are `const fn`, `#[inline]`, `#[must_use]`, delegating to the underlying `u64` saturating operations. - Add unit tests covering normal operation, boundary values, and overflow/underflow saturation.
Ref #6793
ACKs for top commit: Kixunil: ACK ab56fd63b79c22966bc35edda724d8fdbadc6860 Abeeujah: ACK ab56fd63b79c22966bc35edda724d8fdbadc6860 apoelstra: ACK ab56fd63b79c22966bc35edda724d8fdbadc6860; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference✓ Names security-relevant behavior explicitly
Why it was queued
memory safetymerge-commit duplicate discount
AI analysis · Informational 18/100
This commit adds new safe arithmetic helpers to the FeeRate type in the rust-bitcoin library. These methods multiply, add, or subtract fee rates while capping the result at the maximum or minimum allowed value, preventing accidental numeric overflow. It is a routine feature addition with no security fix or vulnerability indicator.
Security candidateMerge rust-bitcoin/rust-bitcoin#6941: consensus_encoding: do not drop encoders with an empty first chunkby Andrew Poelstra · b3789f8b · Oct 11, 2026 · 6 filesMessage 91 · StrongModerate 64Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#6941: consensus_encoding: do not drop encoders with an empty first chunk
`IterEncoder::new` skips leading encoders by testing `current_chunk().is_empty()` on a freshly constructed encoder and discarding it if so. But `Encoder::current_chunk` is explicitly documented as "May return an empty slice", so an encoder is allowed to be:
The fix advances each encoder until it yields a non-empty chunk or reports `Finished`, rather than judging it by its first chunk. `advance()` already handled this correctly; it is moved onto the same helper so the two paths can't drift apart, and two comments claiming a non-existent "chunk must be non-empty" trait contract are removed.
ACKs for top commit: apoelstra: ACK 8eefb6abafb9cce2080e51c0eb94cbcbcb95a1e4; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
consensus or confidential-proof validationmerge-commit duplicate discount
AI analysis · Moderate 64/100
This commit fixes a bug in the way the rust-bitcoin library chains together small data encoders. Previously, if an encoder's first output chunk was empty but it later produced real bytes, the library would silently throw the encoder away and lose those bytes. The fix makes the library advance each encoder past any initial empty chunks before deciding whether to keep it. This could change the bytes produced when encoding certain Bitcoin network messages or other composite data structures.
Security candidateMerge rust-bitcoin/rust-bitcoin#2435: Enforce `$source: std::error::Error` in `write_err`by Andrew Poelstra · 3bb3585a · Oct 11, 2026 · 4 filesMessage 100 · StrongLow 31Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#2435: Enforce `$source: std::error::Error` in `write_err`
71eea3d7f3c15ecae1360ded73c9710aec563e30 Enforce `$source: std::error::Error` in `write_err` (Martin Habovstiak)
Pull request description:
Sometimes people accidentally use `write_err` where `write` should be used or forget to implement `std::error::Error` for their types. So far we've relied mainly on review to catch these but it's difficult to see them and as this commit demonstrates one case slipped in #1400.
This commit add a check to `write_err` that enforces implementation of `std::error::Error + 'static` on types passed to `write_err` as `$source`. This doesn't trigger on `no_std` builds but does in `std` builds. While this would normally be weird in this case it's OK because `std` is tested.
There are still cases where this check is a false positive so the original macro is preserved as `write_err_unchecked_source`.
This change simultaneusly fixes the horrible bug introduced in #1400 which would cause infinite recursion when displaying error on `no_std`.
ACKs for top commit: apoelstra: ACK 71eea3d7f3c15ecae1360ded73c9710aec563e30; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This commit tightens a Rust helper macro used to format error messages so it only accepts real error types. It also fixes a bug where, in builds without the standard library, displaying certain errors would call itself forever (infinite recursion) and crash the program. Most of the changes are defensive compile-time checks; the actual recursion bug is mentioned in the pull request description but the diff only shows the macro change, not the original buggy code path.
AI review queuedMerge rust-bitcoin/rust-bitcoin#7008: primitives: Improve unit testsby Andrew Poelstra · f90bd5b5 · Oct 10, 2026 · 4 filesMessage 91 · StrongInformational 15Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7008: primitives: Improve unit tests
63a578061b8167d2238f4d58f611a10551a06744 primitives: Remove redundant alloc gates in tests (Jamil Lambert, PhD) 4aff57b93625a91766647ec2f4ea59cfeaab6e34 primitives: Fix typo in OutPoint serde test names (Jamil Lambert, PhD) 1520aba850184e11324f3a999e89996beb2c10ed primitives: Test lower hex for all hash types (Jamil Lambert, PhD) 9302d14aecc86204159e859c57ebcc071e59be17 primitives: Test as_byte_array for all hash types (Jamil Lambert, PhD) f835b0e9d5879a1ce73e7e6d07cc5465b48fc075 primitives: Rename witness test module to tests (Jamil Lambert, PhD) d3a0d0edc99d55b1da1eab8aabced5e870321adb primitives: Simplify Merkle root test assertions (Jamil Lambert, PhD) 3a71e1cf7ca0593c50c5eafea108f3cba43d90aa primitives: Remove duplicate Merkle root test code (Jamil Lambert, PhD)
Pull request description:
Several unit tests are inconsistent with the rest of the crate, hard to follow or could be improved:
- `merkle_root_batched` tests against a pasted copy of the default Merkle root code, replace it with `assert_roots_match` cases that test against the real code. - Some Merkle tests use `is_some`/`unwrap` with messages that repeat the test name, compare the option directly. - The witness test module is called `test` not `tests` like the others. - The hash byte array macro only checks `as_byte_array` when given an extra argument that is then ignored, always check it. - The lower hex tests miss `Txid` and `Wtxid`. - Two test names misspell "roundtrips" - Items are gated on `alloc` inside test modules that only exist with `alloc`.
Fix each of these in its own commit. Test changes only.
ACKs for top commit: apoelstra: ACK 63a578061b8167d2238f4d58f611a10551a06744; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This commit only improves and cleans up unit tests in the rust-bitcoin library. It renames misspelled test functions, removes unnecessary feature gates, adds missing test cases for hash types, simplifies Merkle root test assertions, and renames a test module for consistency. There are no changes to production code, so this does not fix or introduce any security vulnerability.
Lower-priorityMerge rust-bitcoin/rust-bitcoin#6995: primitives: Export WitnessMerkleNode encoder and decoder typesby Andrew Poelstra · ca4d1789 · Oct 10, 2026 · 4 filesMessage 91 · StrongTriage 0Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#6995: primitives: Export WitnessMerkleNode encoder and decoder types
`WitnessMerkleNode` implements `Encode` and `Decode`, but its encoder, decoder and decoder error types aren't exported, so users get types they can't name. The `TxMerkleNode` equivalents are exported from `merkle_tree`.
Export them from `primitives` the same way, add them to the API tests and re-export them from `bitcoin`.
ACKs for top commit: Abeeujah: ACK 3a9bd99c87009d20a93ad2d82479be4e0ec56f1f tcharding: ACK 3a9bd99c87009d20a93ad2d82479be4e0ec56f1f apoelstra: ACK 3a9bd99c87009d20a93ad2d82479be4e0ec56f1f; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
AI review queuedMerge rust-bitcoin/rust-bitcoin#7007: primitives: Add track_caller to panicking functionsby Andrew Poelstra · 0ef79473 · Oct 10, 2026 · 3 filesMessage 91 · StrongInformational 21Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7007: primitives: Add track_caller to panicking functions
Policy is to use `#[track_caller]` on functions that panic on invalid arguments. `ScriptBuf::reserve` and `reserve_exact` document a panic but don't have the attribute, and `with_capacity` has the same panic undocumented. `Witness` stores each element's offset as a `u32`, so `from_slice` and `push` panic, undocumented, if an element would start more than `u32::MAX` bytes in.
Remove two stale comments.
Add the attribute and document the panics.
Part of #5518
ACKs for top commit: Kixunil: ACK 83ff3989a412984114917e7cb278665b89a761d4 apoelstra: ACK 83ff3989a412984114917e7cb278665b89a761d4; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This change is a code-quality and documentation improvement, not a security fix. It adds Rust's #[track_caller] attribute to several functions that can panic when given extremely large inputs, and documents those panic conditions. The panics themselves already existed; the patch only makes error messages more accurate and documents behavior. There is no vulnerability being fixed.
AI review queuedMerge rust-bitcoin/rust-bitcoin#7012: units: add track_caller to panic opsby Andrew Poelstra · 4369bd34 · Oct 10, 2026 · 6 filesMessage 100 · StrongInformational 19Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7012: units: add track_caller to panic ops
48fac6da7f6d4186083496cec217eee17ac06124 units: add track_caller to panic ops (satsfy (Renato Britto)) 9a6cbd9a33b16726e482931fbe4bbf7a603aebd6 units: propagate fn attrs in op macro (satsfy (Renato Britto))
Pull request description:
Part of https://github.com/rust-bitcoin/rust-bitcoin/issues/5518
Policy is to use `#[track_caller]` on functions that panic on invalid arguments. In units, there is arithmetic on block heights, weights, fee rates, etc. A panic links code inside units instead of at the caller mistake. The macro now copies fn attributes onto the impls it generates, so that `&a - &b` also links properly.
ACKs for top commit: apoelstra: ACK 48fac6da7f6d4186083496cec217eee17ac06124; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This change adds Rust's #[track_caller] annotation to arithmetic operations in the rust-bitcoin 'units' crate that can panic on invalid input (for example, subtracting a larger number from a smaller one). This annotation does not change whether or when panics happen; it only improves the quality of panic error messages by pointing to the caller's code instead of deep inside the library. It is a debugging/diagnostics improvement, not a security fix.
Security candidateMerge rust-bitcoin/rust-bitcoin#7013: Manual weekly update to rustc (to nightly-2026-10-03) on masterby Andrew Poelstra · 8922579c · Oct 9, 2026 · 4 filesMessage 91 · StrongInformational 15Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7013: Manual weekly update to rustc (to nightly-2026-10-03) on master
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This commit is a routine weekly update of the Rust compiler version used by the project, plus three tiny documentation fixes that add correct backtick formatting around type names. There is no change to program logic, no bug fix, and no security relevance.
AI review queuedMerge rust-bitcoin/rust-bitcoin#7001: units: fix serialization of optional amountsby Andrew Poelstra · 6585c779 · Oct 9, 2026 · 3 filesMessage 91 · StrongLow 48Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7001: units: fix serialization of optional amounts
cfa43314a5616e920201cb1c0d40403cff27a155 units: check Optional amount round trip with bincode (jrakibi) 4b4208ceaf5396b85bfb40e6ae8730e5b0b1a19d units: fix serialization of optional amounts (jrakibi)
Pull request description:
Serializing/deserializing an optional amount using `as_btc::opt` or `as_str::opt` does not round trip with bincode.
The issue is in serialization: we directly serialize the inner value instead of `Some` variant. So deserialization can fail or return incorrect value.
The first commit fixes the issue by using `serialize_some`, and updates serde_bincode to include the two missing bytes that encode `Some` (for unsigned_opt_as_btc and signed_opt_as_btc)
The second commit adds tests.
Found while auditing `units` crate with Codex. This may already be tracked in Loupe. I don’t have access to check.
ACKs for top commit: Abeeujah: ACK cfa43314a5616e920201cb1c0d40403cff27a155 apoelstra: ACK cfa43314a5616e920201cb1c0d40403cff27a155; successfully ran local tests; great catch!
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This commit fixes a bug in how optional Bitcoin amounts are serialized when using certain compact binary formats (like bincode). The code was writing the raw inner value for a Some(amount) instead of properly marking it as 'Some'. This could cause deserialization to fail or produce wrong values when round-tripping optional amounts. The fix changes the serializer to use the standard 'serialize_some' method and adds tests to confirm round-tripping now works.
AI review queuedMerge rust-bitcoin/rust-bitcoin#7003: Fix creusot cache by running on push to `master`by Andrew Poelstra · bc026dc2 · Oct 9, 2026 · 1 fileMessage 100 · StrongInformational 15Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7003: Fix creusot cache by running on push to `master`
8f9a6dec87f84c5b125b144b8d6144d8437af5cb Fix creusot cache by running on push to `master` (Martin Habovstiak)
Pull request description:
According to the [documentation], PRs don't get to share their cache for security reasons which severely limits its usefulness to us. They can however restore the cache of the base branch which solves the issue but to have any cache on base branch we have to run Creusot on push.
This commit enables the cache on pushes to `master`. It is expected that this will require a full re-build but all subsequent builds which do not modify the dependencies should get faster.
Adding P-high label because not having this wastes a lot of time on every PR and fills our cache storage.
ACKs for top commit: apoelstra: ACK 8f9a6dec87f84c5b125b144b8d6144d8437af5cb; successfully ran local tests jamillambert: ACK 8f9a6dec87f84c5b125b144b8d6144d8437af5cb
This change simply tells GitHub to run the Creusot verification workflow whenever code is pushed to the master branch, in addition to running it on pull requests. The goal is to populate a build cache for the master branch so that future pull requests can reuse it and run faster. There is no change to the actual Bitcoin library code, no bug fix, and no security patch.
Lower-priorityMerge rust-bitcoin/rust-bitcoin#7002: internals: Restore T: Eq in ArrayVec Eq implby Andrew Poelstra · 660e263a · Oct 9, 2026 · 1 fileMessage 100 · StrongTriage 0Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7002: internals: Restore T: Eq in ArrayVec Eq impl
8ca009407c18bbc1b81a8c417de20034a43540e4 internals: Restore T: Eq in ArrayVec Eq impl (jrakibi)
Pull request description:
The bound was changed from `T: Eq` to `T: PartialEq` in #6942. This allows ArrayVec to implement Eq for types such as f64, even though f64 does not implement Eq (because NaN != NaN).
Change the bound back to T: Eq
Follow up to https://github.com/rust-bitcoin/rust-bitcoin/pull/6942#discussion_r4216067635
ACKs for top commit: Kixunil: ACK 8ca009407c18bbc1b81a8c417de20034a43540e4 apoelstra: ACK 8ca009407c18bbc1b81a8c417de20034a43540e4; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Security candidateMerge rust-bitcoin/rust-bitcoin#7000: internals: add track_caller to U256 opsby Andrew Poelstra · 0870db2e · Oct 9, 2026 · 2 filesMessage 86 · StrongInformational 20Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#7000: internals: add track_caller to U256 ops
276712155d4467198fa7797afbd530467a8ea2e3 internals: add track_caller to U256 ops (satsfy (Renato Britto))
Pull request description:
Part of https://github.com/rust-bitcoin/rust-bitcoin/issues/5518#issuecomment-5981738769
In debug builds, internals Add, Sub and Mul on U256 panic on overflow, Div and Rem panic on a zero divisor.
The panics linked code inside internals instead of at the caller's mistake, and the policy document asks for `track_caller` on functions that panic on invalid arguments.
ACKs for top commit: Kixunil: ACK 276712155d4467198fa7797afbd530467a8ea2e3
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Links an issue, advisory, or supporting reference✓ Names security-relevant behavior explicitly
Why it was queued
memory safetymerge-commit duplicate discount
AI analysis · Informational 20/100
This change adds Rust's #[track_caller] attribute to several internal math and array helper functions. It does not fix a bug or change any calculation; it only improves panic error messages so that if an overflow or division-by-zero happens in testing, the error points to the caller's code rather than deep inside the library. There is no security vulnerability being patched.
Lower-priorityMerge rust-bitcoin/rust-bitcoin#6999: units: Make parse_int helper functions privateby Andrew Poelstra · d9309fcd · Oct 9, 2026 · 2 filesMessage 91 · StrongTriage 0Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#6999: units: Make parse_int helper functions private
01bedbb8835f977f8d198440a2d87528ac97ed8b units: Make parse_int helper functions private (Tobin C. Harding)
Pull request description:
These two functions are not needed publicly. They are helper functions.
Make the two functions `pub(crate)`.
ACKs for top commit: satsfy: tACK 01bedbb8835f977f8d198440a2d87528ac97ed8b Kixunil: ACK 01bedbb8835f977f8d198440a2d87528ac97ed8b apoelstra: ACK 01bedbb8835f977f8d198440a2d87528ac97ed8b; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Security candidateMerge rust-bitcoin/rust-bitcoin#6998: Fix regtest parameters to match Bitcoin Coreby Andrew Poelstra · 621a3900 · Oct 9, 2026 · 1 fileMessage 91 · StrongInformational 19Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#6998: Fix regtest parameters to match Bitcoin Core
8c0930daac175ef912c533186c2108e1d5d0da67 params: Align regtest values with Bitcoin Core (moisesPomilio)
Pull request description:
Updates the regtest parameters to match Bitcoin Core's current regtest configuration.
* Sets BIP34, BIP65, and BIP66 activation heights to 1. * Changes the regtest difficulty adjustment timespan from 2 weeks to 1 day.
Fixes #6997
ACKs for top commit: apoelstra: ACK 8c0930daac175ef912c533186c2108e1d5d0da67; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
proof-of-work or emission validationmerge-commit duplicate discount
AI analysis · Informational 19/100
This commit updates the built-in 'regtest' network parameters in the rust-bitcoin library so they match Bitcoin Core's current regtest defaults. Regtest is a private test network used only for local development and automated testing, not for real money. The changes make several Bitcoin protocol rules active from block 1 and shorten the difficulty adjustment window from two weeks to one day. There is no indication this fixes a security vulnerability; it is a compatibility/maintenance fix.
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
This commit is a routine release-tracking merge that bumps the version of the `bitcoin-internals` crate from 0.7.0 to 0.8.0 and updates all dependent crates' version requirements and lockfiles. It does not contain any code changes, bug fixes, or security patches itself. The changelog notes that other pull requests included in this release address items such as sanitizing serde size hints and fixing work/target conversions, but those changes are not part of this merge commit's diff.
✓ Descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
cryptography-sensitive path
AI analysis · Informational 15/100
This commit only fixes documentation formatting by adding Rustdoc backtick link syntax to comments. It changes no executable code, no APIs, and no behavior. There is no security relevance.
The relative `IncompatibleHeightError` is defined as "Tried to satisfy a lock-by-time locktime...", but its message says "tried to satisfy a lock-by-height locktime...". `IncompatibleTimeError` has the same inversion.
Fix the lock type and the value used to check it.
ACKs for top commit: apoelstra: ACK 1be2848cb7d08c9cc6f83f0d372b462b55b5ca2f; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Security candidateMerge rust-bitcoin/rust-bitcoin#6963: Clean up Pedantic Clippy Lintsby Andrew Poelstra · 82d60e08 · Oct 8, 2026 · 21 filesMessage 91 · StrongInformational 15Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#6963: Clean up Pedantic Clippy Lints
38de85291f7b17e909590407b8d9510d4e374b91 bitcoin: Fix pedantic clippy lints (Abeeujah) 463a5c1f13c86281445fbfa8314e41aac9946816 addresses: Use Option::map_or for `rfind` (Abeeujah) a86f9029c889741bbfb9671f872cb7287afffbfb p2p: Use Option Combinators (Abeeujah) 74c132c866f614a4aa10edf1ae4a91e613d0c480 primitives: Fix pedantic clippy lints (Abeeujah) c61c04fc892432fcfabbff0ee04aa5e2c01b3f13 clippy: Use Option combinators (Abeeujah) c1e940cd4c629406d4e116fc8de01d3f3097f07a Drop deprecated from_iter_instead_of_collect lint (Abeeujah)
Pull request description:
This PR does a sweep across all crates where fixing the triggered clippy lints would provide some value.
ACKs for top commit: apoelstra: ACK 38de85291f7b17e909590407b8d9510d4e374b91; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
cryptography-sensitive pathsigning or wallet pathmerge-commit duplicate discount
AI analysis · Informational 15/100
This is a routine code-cleanup pull request that fixes style warnings reported by the Clippy linter. It replaces verbose match/if-let blocks with shorter Option combinators, removes an obsolete lint rule, and uses explicit integer conversion helpers. There is no security bug being fixed and no behavior change intended.
AI review queuedFix creusot cache by running on push to `master`by Martin Habovstiak · 8f9a6dec · Oct 8, 2026 · 1 fileMessage 81 · StrongInformational 15Details
Commit message · Martin Habovstiak
Fix creusot cache by running on push to `master`
According to the [documentation], PRs don't get to share their cache for security reasons which severely limits its usefulness to us. They can however restore the cache of the base branch which solves the issue but to have any cache on base branch we have to run Creusot on push.
This commit enables the cache on pushes to `master`. It is expected that this will require a full re-build but all subsequent builds which do not modify the dependencies should get faster.
This commit changes the project's automated GitHub workflow so that a verification tool called Creusot also runs whenever code is pushed to the main 'master' branch. The only goal is to make the workflow's build cache work better for pull requests. There is no change to the actual Bitcoin library code, no bug fix, and no security vulnerability being addressed.
The bound was changed from `T: Eq` to `T: PartialEq` in #6942. This allows ArrayVec to implement Eq for types such as f64, even though f64 does not implement Eq (because NaN != NaN).
Change the bound back to T: Eq
88/100 · StrongMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Links an issue, advisory, or supporting reference
units: check Optional amount round trip with bincode
Add bincode round trip checks to the existing as_btc::opt and as_str::opt tests.
75/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Mentions testing or verification
AI review queuedunits: fix serialization of optional amountsby jrakibi · 4b4208ce · Oct 8, 2026 · 2 filesMessage 80 · StrongModerate 50Details
Commit message · jrakibi
units: fix serialization of optional amounts
Optional amounts do not round trip with bincode because we directly serialize the inner value instead of `Some` variant. Deserialization can therefore fail or return incorrect values.
Use serialize_some in as_btc::opt and as_str::opt.
Also update serde_bincode to include the two missing bytes that encode `Some` (for unsigned_opt_as_btc and signed_opt_as_btc)
80/100 · StrongMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode
Why it was queued
second-pass: broader security terminology
AI analysis · Moderate 50/100
This commit fixes a bug in how optional Bitcoin amounts are serialized when using the bincode format. Previously, an optional amount like `Some(1 BTC)` was written as just the raw number/string, without the marker that says "this value is present." That meant when reading it back, the code could misinterpret the data or fail. The fix uses the standard `serialize_some` method so the `Some` wrapper is preserved. The test data file was also updated to include the two missing bytes that represent `Some`.
d54990ae3c59f01671add359f2e59a9f05bcaac8 primitives: Fix witness version error message (Jamil Lambert, PhD) a8f89a058befc6bb6dbe1eb032fe916920e2107f primitives: Add From<Infallible> for witness error (Jamil Lambert, PhD)
Pull request description:
`InvalidWitnessVersionError` is the only public error that doesn't implement `From<Infallible>` (P-ERROR-INFALLIBLE), and its message says "invalid witness script version".
Add the impl and change the message to "invalid witness version".
ACKs for top commit: apoelstra: ACK d54990ae3c59f01671add359f2e59a9f05bcaac8; successfully ran local tests Abeeujah: ACK d54990ae3c59f01671add359f2e59a9f05bcaac8 tcharding: ACK d54990ae3c59f01671add359f2e59a9f05bcaac8
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Lower-priorityMerge rust-bitcoin/rust-bitcoin#6994: primitives: Test parsing WitnessVersionby Andrew Poelstra · 2e69a9cd · Oct 7, 2026 · 1 fileMessage 91 · StrongTriage 0Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#6994: primitives: Test parsing WitnessVersion
d217a1b113d1673e80b50220adbb3af3df4e3bfc primitives: Test parsing WitnessVersion (Jamil Lambert, PhD)
Pull request description:
There are no unit tests for parsing a `WitnessVersion` from a string or for the error returned by `TryFrom<u8>`, so the range check and error variants are only covered by a doc example.
Add tests for both.
ACKs for top commit: apoelstra: ACK d217a1b113d1673e80b50220adbb3af3df4e3bfc; successfully ran local tests Abeeujah: ACK d217a1b113d1673e80b50220adbb3af3df4e3bfc tcharding: ACK d217a1b113d1673e80b50220adbb3af3df4e3bfc
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
merge-commit duplicate discount
Security candidateMerge rust-bitcoin/rust-bitcoin#6989: units: drop PartialOrd and Ord from CompactTargetby Andrew Poelstra · 7074bfe1 · Oct 7, 2026 · 3 filesMessage 91 · StrongLow 28Details
Commit message · Andrew Poelstra
Merge rust-bitcoin/rust-bitcoin#6989: units: drop PartialOrd and Ord from CompactTarget
78c5fce5a3c46509157637681c0b333143e119bf units: drop PartialOrd and Ord from CompactTarget (satsfy (Renato Britto)) 9162fe34bfe5816ac893ef8d691e0aaf83b06d05 primitives: implement lexicographic Header Ord (satsfy (Renato Britto))
Pull request description:
Close https://github.com/rust-bitcoin/rust-bitcoin/issues/2758 See discussion on https://github.com/rust-bitcoin/rust-bitcoin/pull/2754#issuecomment-2102736580
Conversion between `CompactTarget` and `Target` is lossy both ways. A derived order on the underlying `u32` does not correspond to which target is actually harder. The policy doc says to consider carefully before deriving `PartialOrd` or `Ord` on types where the comparison has no semantic meaning, and this is one instance of. `block::Header` and p2p `HeaderAndShortIds` contain a `CompactTarget`, they get derive removed.
About equality, it stays on the consensus encoded `u32` and that is ok. Simplified the docs for it.
The issue also suggests `ArbitraryOrd` but we removed that some time ago, so nothing takes its place and its up to the user to convert their `CompactTargets` to compare.
Note that the issue and discussion is somewhat old (2024), but it still seems just as relevant, so I jumped directly to PR.
ACKs for top commit: tcharding: ACK 78c5fce5a3c46509157637681c0b333143e119bf apoelstra: ACK 78c5fce5a3c46509157637681c0b333143e119bf; successfully ran local tests
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
consensus or confidential-proof validationmerge-commit duplicate discount
AI analysis · Low 28/100
This change removes automatic numeric ordering from a Bitcoin type called CompactTarget because its underlying u32 value does not reliably reflect which mining target is actually harder. It also replaces the derived ordering of block headers with a clear byte-by-byte (lexicographic) ordering based on how headers are serialized. The patch is a correctness and API-safety improvement rather than a fix for an active exploit, but it prevents future bugs where code might accidentally sort or compare CompactTarget values as if the raw number meant 'harder' or 'easier'.